Studies have shown African-American men (AAs), who have the highest prostate cancer (PC) rate worldwide, also have relatively higher Prostate-Specific Antigen (PSA) levels and more Digital Rectal Exam (DRE) abnormalities that correlate with current or eventual PC. This may be related to higher androgen levels in AAs, or to greater exposure to environmental carcinogens such as dietary heterocyclic amines (HAs). The major HAs, formed mostly in well-done meats, cause colon and mammary cancer and PC in rats. Increased tumor risks for these sites have been associated with eating well-done meat in case-control studies. However, no study has focused on HA exposure in AAs, despite observations that AAs ingest higher quantities of meat and more thoroughly-cooked meat than do whites. This 5- year clinic-based study will investigate the association of dietary mutagen exposure and prediagnostic screening indicators of prostate cancer risk in African-Americans. The study will include 140 cases (with abnormal PSA and/or DRE screening results) and 430 to 460 controls. Targeted solicitation of participants from organizations that serve primarily AAs in or near Oakland, California, will be implemented to gather data from potentially HA-exposed AAs who otherwise would not be likely to obtain early PC screening. This study aims specifically to: (1) estimate dietary HA exposures in AAs; (2) collect PSA and DRE screening results for AAs and use these data to test hypotheses relating higher HA exposures to increased abnormality in PC screening results. To accomplish these aims, AAs will be interviewed about dietary practices and will be asked to provide 12-hour urine specimens and cooked-food samples prior to prostate screening. Follow-up clinical diagnostic data obtained for abnormal PC screenees will be analyzed separately for HA-related associations together with corresponding prediagnostic PC-screening data. Data analyses will include multivariate logistic regression, general additive and linear models, and analysis of variance using packaged computer programs. By clarifying how HA exposure may be associated with prediagnostic PC-risk indicators, this study will help define diet and food preparation interventions to reduce risk of PC in the population at greatest risk.